• Aucun résultat trouvé

The concept of ALARA, which is synonymous with the optimization of

radiation protection has been in use for some time. As a result, there are a number of examples* of optimization applications available in the relevant literature that might be useful as guidance.

Several examples are discussed below and, as needed, some additional comments are included to indicate the important optimization principles involved.

V.I General example

ICRP Publication 37 gives a general example of the optimization of radiation protection in the transport of radioactive materials by road, rail and waterways, and considers both normal and accident conditions. Although the example is not quantitative in nature, it does set out principles for the application of the optimization concept and is reproduced below.

"The transportation of radioactive materials involves regular movements of materials associated with the nuclear fuel cycle (fissile material, fresh reactor fuel, spent reactor fuel, radioactive wastes) and with various applications of radionuclides and radiation sources in

research, industry and medicine. This process takes place over a network of pathways connecting places where such materials are produced and

used. Each movement between two of the above points involves a radiation risk to the workers involved and to the population living along the

transport pathway, as a result of exposure in normal conditions of

transport and of possible exposure in the event of an accident leading to a release of radioactive material. A theoretical example of

optimization, aimed at dealing only with routing of radioactive

materials, will be discussed. It should be noted, however, that the most effective contribution from optimization might be found in its

application to the standard design of containers for relatively small radioactive sources.

"The potential exposure associated with transportation accidents, as well as the actual exposure in normal transport conditions, can be

estimated in terms of detriment. The consideration of accidental exposures and their detriment is beyond the scope of this example,

particularly as far as low-probability accidents involving high doses are concerned; however, this type of consideration could override

optimization of radiation protection. In any case, the protection needed to avoid or mitigate more probable minor accidents leading to low level exposures to radiation might be formally optimized. If so, the attention should be focused not only on the detriment associated with normal

transport operations but also on that resulting from minor accidents during transport.

"In order to assess the detriment to the population due to a given transport operation, it would be necessary: (a) to determine the

corresponding source term, including type and amount of radioactive material being transported, features of transport packages, radiation exposure rate as a function of distance within and around the transport vehicle, likely minor accidental releases (and their probabilities); (b) to develop a dosimetric model allowing for the calculation of the

distribution of dose equivalent to individuals at various locations along the transport pathway (this will depend on the radiation exposure rate around the vehicle, the speed of the vehicle, the physical features of the pathway and other factors); and (c) to apply the above model to the actual distribution of population within a given distance from the road, rail or waterway in order to calculate the collective dose commitment associated with the given transport operation along the particular pathway. This should include the contribution from the exposure of the transport operators, and should apply to normal transport conditions as well as to minor accidents.

"The calculation can conceptually be extended to an entire network of transport pathways connecting the points of production and use of radioactive materials. Thus, the total detriment associated with the full system of radioactive material transport in the country can

theoretically be evaluated. Generally, the transportation between any two points in the system may be achieved through different alternative pathways of transport (air, road, rail or waterway) with which different values of radiation detriment are generally associated owing to the different features of the transport (packaging standards, speed, pathway characteristics) and the different distribution of population. These alternatives generally imply differences in the transport costs due to

the different lengths of the alternative pathways, as well as to the type of transport means and the pathway features. If the detriment and the cost associated with each alternative transport network and system considered could be formulated, a theoretical optimization can be

performed using the methods described in Section B." (Note: Section B referred to here is Section B of Reference 1.)

"The above example illustrates a particular case in which the

optimization of protection is unlikely to be a significant factor in the planning of transport operations, as distinct from the design of

packaging. Other factors, e.g. physical security, could play an important part in planning the operations themselves. The example, nevertheless, sets out the principles for the application of the optimization concept."

V.2 Examples involving normal transport

The proceedings of PATRAM '80 held in Berlin (West) and of PATRAM '83 held in New Orleans contain & few papers on optimization in radiation protection. At the Berlin symposium Murphy et al presented a paper on(2) the application of ALARA principles to the shipment of spent fuel and

Sutherland gave a paper on an ALARA assessment of spent fuel and nuclear(3) waste transport systems.

Murphy considered various options in transport equipment design.(2) Radiation doses were calculated and converted to costs. The alternative equipment designs were also costed and the following conclusions presented:

"Public exposure to radiation from spent fuel shipments is low. Occupational and transport worker doses account for the majority of the total dose. Any efforts to reduce dose should be concentrated in this area. Practices that affect the age of spent fuel in shipment and the number of times the fuel must be shipped prior to disposal have the largest impact on accumulated dose. A policy to encourage a 5-year spent fuel cooling period prior to shipment coupled with appropriate cask redesign to accommodate larger loads would be consistent with ALARA principles. Rerouting shipments to avoid high

population density areas will not reduce total shipment dose due to a corresponding increase in occupational dose."

Sutherland did not perform a complete cost-benefit assessment per(3) se, but provided data which could be used in such an assessment. Parametric

assessments were performed for cask designs depending upon the age of spent fuel, allowable surface dose rates, and cask radiation shield configurations, and these data were then used to determine the lifetime transportation costs for each configuration. It was then concluded that "(1) some low dose rate cask designs may be oversize or overweight if lead or steel provide the gamma shielding, (2) designing for 5 instead of 10 mrem/h increases the cask weight by about 10% and for 2 mrem/h adds another 10% or more to the weight, and (3) the increased weight of the low dose rate casks results in an estimated $1 to

$2 million increase in the cask lifetime transportation costs."

At PATRAM '83 there were 8 papers given on risk analysis techniques and 8 papers presented on risk analysis. Optimization is the subject of only one of these papers, that of Palacios and Menossi on optimization of radiation(4) in the transport of radioisotopes. In the study, approximately 7300 shipments per year of radioisotopes (principally Mo/Tc-99m generators and 1-131

compounds); in six types of containers; in special vehicles, commercial planes and long-distance buses; transported throughout Argentina were considered. A cost-benefit analysis, using a value of a = US$ 10,000 showed that increased shielding on packages would be required to obtain an optimized situation.

Additional shield thickness of from 0.9 to 1.5 cm of lead with 7% antimony could be justified, reducing the radiation levels l m from the package surface from approximately 25 vtSv/h to 5 v.Sv/h for the transport of Mo/Tc-99m and to 2.5 p.Sv/h for that of 1-131. The paper concludes by stating:

"The values obtained for Argentina cannot be easily extrapolated

internationally due to the fact that protection costs and detriment vary from one country to another. However, the cost of protection, X, is highly affected by the cost of the lead used as a shield. Since the cost of lead follows the international metal prices, substantial variations should not be expected among the various countries. On the other hand, the value of a (US$ 10,000/man Sv) adopted by the Argentine authority might be too high for its international acceptance. If a value liable to be more easily acceptable internationally were adopted, such as

US$ 5,000/man Sv, the optimized level would remain in about 10 v-Sv/h at l m from the packaging."

Ringot provides useful information in his paper on radiation

exposure of transport personnel. These data, which are summarized in Annex IV, provide potential input for optimization studies.

V.3 Examples involving accident situations and risk assessment

Hubert and Pages in their paper on optimization in transport describe the stages involved in their risk assessment as follows:

"The first step includes a description of the logistics of the

transportation (quantities, number of shipments, modes, shipping routes), the description of the reference shipment (type of vehicle, type of

package, amount shipped, safety devices, etc.).

"The analysis of the shipped material also takes place at this step. It involves the description of its physical and chemical

properties, and of the nature of the hazards associated with the loss of the integrity of the package. The hazard can pertain to chemical or radiological toxicity or both. A release can lead to acute or delayed effects such as latent cancer fatalities and genetic effects. In the case of a radioactive material that is non-dispersible, one must consider the external irradiation due to the loss of shielding. In the case of an atmospheric release of dispersible material several pathways will have to be modelled: external irradiation from ground deposits and cloud,

internal contamination through inhalation or ingestion of contaminated food. In some cases the risk of criticality should also be considered.

All these cases involve quite different time scales and geographic

areas. The preliminary analysis performed at this step often allows one to neglect some aspects of the risk. When performing an ALARA analysis, the modelling should be performed with special regards to the alternative safety options at hand in the optimization process. If there is no

alternative the model can be simple, if there is one then an appropriate parameter must be present to take it into account. This attitude is often disagreed with on the ground that important parameters are neglected whereas some others of minor importance are taken into

account. But that inconsistency disappears when optimization analyses are performed, for the models are used in a comparative manner and not for absolute assessments. At least the general environment must be detailed, that is the actual traffic conditions, the demographic parameters and the weather condition probabilities along the selected routes.

"The second step aims to describe the accident environment - with a view to enable assessing probabilities of events and eventually of the

various accident scenarios. At the end of this step the different thermal and mechanical loads which can be applied to the package must have been identified, and the probability distribution of their values estimated.

"In the third step, the package behaviour under the accidental

stresses is examined. The probability of a loss of protection as well as the magnitude of a release are associated to the intensity of the loads that can be encountered.

"The last step leads to the assessment of health consequences, with the use of transfer models and demographical parameters. The assessment is generally performed for a first reference option. Then risk is

assessed for the other alternatives in order to quantify their

effectiveness. Altogether with the assessment of the cost, an ALARA optimization can be done with these data."

The point is made that such probabilistic risk assessments for some systems can be very complex but others may be quite simple.

Other papers concerned with accident conditions include "Cost effectiveness of safety measures applying to hexafluoride transport in France" and a study concerned with transport through the Mont Blanc

/ Q \

Tunnel . Because of the possible nonavailability of Reference 8, an abstract of this work follows:

"Modifying the regulation for small radioactive package transit through the Mont Blanc Tunnel. An assessment of the health and economic impact."

"The Mont Blanc Tunnel is a 12 km long tunnel, under the highest mountain in Europe, and is one of the most important routes between France and Italy. Hazardous materials are subjected there to much more stringent regulations than is required on plain highways by international regulatory convention (ADR or IAEA). Presently the transportation of

small amount of dispersable radioactive material (i.e. whose

transportation is allowed in "Type A" packages under a given threshold in activity) is permitted for a truck content which is only one third of the activity limit - A - that is applicable to a Type A package. This

probabilistic study aims to assess the implications of a move towards the application of the ADR convention.

"The implications of an accident under the tunnel have first been investigated. Health consequences might be linked to the traffic accident itself or to the release of radioactive materials. In the latter case immediate deaths and late effects of low doses (delayed causes and genetic effects) must be distinguished. There are also three kinds of direct economical impacts: the damage to the tunnel and

vehicles, the cost of traffic interruption, and the cost of decontamination.

"Some of these impacts would increase with a change in regulation, but others would decrease and still others would remain unchanged.

Allowing for example three times more activity in a vehicle means that the number of shipments is divided by three, and so is the number of accidents. Thus the traffic victims and the damage to the tunnel and vehicles would decrease by a factor of three. A preliminary analysis showed that the cost of decontamination and traffic interruption would also decrease, since these are less than proportional to transported activity. The delayed effects of radiation being linear with the dose, and the dose being linear to the activity, would not be modified. On the other hand, immediate effects are most likely to increase.

"The decisional process implies knowing the answers to these two questions. Are the new levels of risk acceptable? Is the measure cost effective?

"A conservative approach has been used to answer the first question, with the transport of a total immediate release by air along the tunnel being assessed using a compartment model, and a puff model in the first compartment. The accident individual dose to the public would be about 6 10 Sv in the most contaminated part of the tunnel, instead of 2 in_2 the case of the present activity limit. These figures illustrate that the conditions of dispersion within the tunnel are quite close to the assumed hypothesis for deriving this A? limit in the IAEA guidelines.

"Immediate death can follow exposure in a tiny volume downwind the source. In the most severe compartment lethality can be observed up to 2.5m downwind, but within a 10 cm radius. Such a volume is about 5 times bigger (3 ) than it would be for the 1/3 A limit, and so is3/2 the probability to encounter such effects for a given traffic. It is very likely that the "acceptability" of the risk would be the same in the two cases.

"A risk benefit analysis is then conducted. But as all

nonradiological risks are decreasing and as they are more important than radiological risk, the results are that such an option both saves money and reduces risk. The point is rather to study the transfer from

nonradiological to radiological risk. Presently a test programme on the behaviour under accidental stresses of these packages is carried out in the CEA to improve the accuracy of this probabilistic risk assessment and future safety studies."

References (for Appendix V)

1. ICRP Publication 37, Cost Benefit Analysis in the Optimization of Radiation Protection, 1982.

2. D.W. Murphy, J. Greenberg, L.W. Brakenbush, R.A. Burnett, J.R. Lewis and W.B. Andrews, "Application of ALARA Principles to Shipment of Spent

Nuclear Fuel", PATRAM '80, Proceedings of the 6th International Symposium on Packaging and Transportation of Radioactive Materials, p. 206, 1980.

3. S.H. Sutherland, "An ALARA Assessment of Spent Fuel and Nuclear Waste Transportation Systems", PATRAM '80, Proceedings of the 6th International Symposium on Packaging and Transportation of Radioactive Materials, p.

211, 1980.

4. E. Palacios and C.A. Menossi, "Optimization of Radiation Protection in the Transportation of Radioisotopes", PATRAM '83, Proceedings of the 7th International Symposium on Packaging and Transportation of Radioactive Materials, p. 192, 1983.

5. C. Ringot, "Study on the Radiation Exposure of Transport Personnel", CEA Report, IAEA Research Agreement No. 2792/R/CF, March 1983.

6. P. Hubert and P. Pages, "The application of ALARA principles to the safety of radioactive materials transportation", Proc. Radiation Protection Optimization in Nuclear Fuel Facilities, Luxembourg 1983.

7. P. Hubert, P. Pages and B. Augin, "Cost effectiveness safety measures applying to uranium hexafluoride transportation in France", PATRAM 1983, New Orleans, May 1983.

8. P. Pages, P. Hubert, P. Gilles, J. Hamard, C. Ringot, and E. Tomochevsky,

"Modifying the regulation for small radioactive package transit through the Mont Blanc Tunnel - An assessment of health and economic impact", CEA Report.

List of Participants, Consultants, Commentors. and Contributors

Consultants Meeting, 3-7 December 1984 Japan - A. Kasai

United Kingdom - K.B. Shaw International Atomic

Energy Agency - R.B. Pope (Scientific Secretary)

Argentina

-Commentors on First Draft. Jan.-May 1985 A.L. Biaggio Korea, Republic of J.W. Ha

Technical Committee Meeting (TC-555). 17-21 June 1985 Belgium

United States of America

Documents relatifs